Multimedia applications are becoming an integral part in the design of wireline communication systems. These applications focus on the transmission of images, data, and speech over the same communication channel. A particular concern in multimedia applications is the transmission of images. An image comprises a two-dimensional signal that represents the relative luminances of objects in a scene. A scene may comprise, for example, a photograph, a video image, an X-ray image, or a radar image for a weather forecast. Two important parameters are the quality and the intelligibility of the image resulting from the transmission of the image data.
A typical image consists of a matrix of 512.times.512 pixels. For uncolored images, each pixel is described by a gray level and uses 8 bits of memory to store this information. The result is that the entire image occupies approximately 2 Mbits of memory. For a colored image, each pixel uses 24 bits and results in a image occupying approximately 6 Mbits of memory. Thus, a full-detail color image usually requires a lot of memory, bandwidth, and power consumption.
As a natural result of advances in technology, attention is being given to the transmission of multimedia signals via radio frequency communication. In the near future, this communication vehicle will become an important part of the service requirements for radio frequency communication systems. However, unlike wireline communication systems, radio frequency systems tend to be spectrally limited. In addition, service costs, in terms of air time charges, are significantly higher for the consumer. Images usually occupy large quantities of device memory and the transmission of image signals take a correspondingly long time. As such, it becomes quite expensive to transmit images via radio frequency systems such as mobile cellular radio networks. These resources need not be wasted if high resolution image transmission is not required. Therefore, a system is needed whereby the air time necessary to transmit images via radio frequency systems can be minimized in order to minimize costs to the user.